davidradl commented on code in PR #26962:
URL: https://github.com/apache/flink/pull/26962#discussion_r2322481809
##########
flink-table/flink-table-runtime/src/main/java/org/apache/flink/table/runtime/operators/window/groupwindow/triggers/ProcessingTimeTriggers.java:
##########
@@ -109,10 +109,7 @@ public boolean onProcessingTime(long time, W window)
throws Exception {
ReducingState<Long> nextFiring =
ctx.getPartitionedState(nextFiringStateDesc);
Long timer = nextFiring.get();
if (timer != null && timer == time) {
- long newTimer = time + interval;
- ctx.registerProcessingTimeTimer(newTimer);
Review Comment:
I am curious about the reasoning why this
ctx.registerProcessingTimeTimer(newTimer); can be removed. The effect would
appear to be that instead of a chain of timers, there will now will just be one
for the window. The comments in this file say " Trigger every a given
interval, the first trigger time is interval after the first element in
the pane." . The first trigger time implies there will be subsequent ones.
I am not sure how this change is scoped to the early/late firing scenarios
mentioned in the PR conversation.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
